23 papers
A parallel-in-time Newton's method-based ODE solver
Casian Iacob, Hassan Razavi, Simo Särkkä
In this article, we introduce a novel parallel-in-time solver for nonlinear ordinary differential equations (ODEs). We state the numerical solution of an ODE as a root-finding prob…
A Bayesian Filtering Approach for Learning Lagrangian Dynamics from Noisy Measurements
Kundan Kumar, Shreya Das, Simo Särkkä
This paper proposes a Bayesian filtering-based approach for learning the dynamics of a physical system from partial, noisy measurements. We model the system dynamics using a Lagran…
Scalable Gaussian Processes for Integrated and Overlapping Measurements Via Augmented State Space Models
Ryan A. Rubenzahl, Soichiro Hattori, Simo Särkkä +4
Astronomical measurements are often integrated over finite exposures, which can obscure latent variability on comparable timescales. Correctly accounting for exposure integration w…
Integrating Lagrangian Neural Networks into the Dyna Framework for Reinforcement Learning
Shreya Das, Kundan Kumar, Muhammad Iqbal +4
Model-based reinforcement learning (MBRL) is sample-efficient but depends on the accuracy of the learned dynamics, which are often modeled using black-box methods that do not adher…
Online Bayesian Experimental Design for Partially Observed Dynamical Systems
Sara Pérez-Vieites, Sahel Iqbal, Simo Särkkä +1
Bayesian experimental design (BED) provides a principled framework for optimizing data collection by choosing experiments that are maximally informative about unknown parameters. H…
Dual-Level Models for Physics-Informed Multi-Step Time Series Forecasting
Mahdi Nasiri, Johanna Kortelainen, Simo Särkkä
This paper develops an approach for multi-step forecasting of dynamical systems by integrating probabilistic input forecasting with physics-informed output prediction. Accurate mul…